is N+T req'd for P91 induction bend? 2

[davefitz]

When I read the Mannesman guidebook for fabricating components of P91, it states that all hot formig operations must be followed by a full normalized + tempering N+T treatment of the entire piece. This is to ensure that any sections that may have austenized or become overtempered by exposure to a temperature range 1500-1800F have their crystal structure repaired.

I have just been advised that there are many recent combined cycle sites which had their P91 HP main steam line bends fabricated using an induction heating bending machine, and these bent pieces did not later undergo a N+T . Does this mean we can expect a rash of premature P91 failures in the near future? And why doen't ASME code require the N+T?

 

[stanweld]

Unfortunately, premature failures should be expected. Premature tube failures have been reported on tubes that were tempered at 1500F by the tube manufacturer. I have seen some bending procedures that indicated heating between 1500 F and 1650 F without subsequent N&T. These procedures were of course rejected by me and N&T following bending was required. How many hot bends were made without N&T and installed are unknown but I suspect we should find out after about 30,000 hrs operation.

 

[metengr]

I would like to add that we have had several HRSG's recently commissioned in the State of california by our sister power company, and our induction bending was performed by BendTec. They are only one of the few in the US that can properly induction bend power piping and typically recommend a normalize and temper for P91 material. Most boiler OEM's also have expertise regarding handling of P91 material.

In this business, the customer must beware and understand material limitations. I believe the reason the Code is silent on this issue is it hasn't been a concern. The Code deals with design, material selection and joining to result in a safe component. Bending is one of those detailed fabrication steps that until serious problems develop or enough users complain about it, nothing will be said or done about it.

As far as a rash of P91 failures, the only way we will know will be operating experience. My feeling is that P91 used on Power Boilers is probably not going to be a problem because moist utilities have the in-house technical expertise or use experienced suppliers to avoid mishandling of P91 piping. I would believe that smaller IPP's are probably left exposed and unless they have someone like stanweld, you or myself on staff - they will be in a world of hurt.

 

[SMF1964]

As I've opined in the past, the problem with relying on a tempered martensite is that any thermal excursion (such as induction heating) that elevates the temperature above the original tempering temperature will alter the mechanical properties of the P91. I disagree with the statement that P91 used on Power Boilers is probably not going to be a problem because a number of these new constructions are turnkey items and out of the hands of the technical expertise at the utilities, relying instead on the 'experienced suppliers' and erectors to do it right. The entire harp may arrive heat treated properly, but when you weld it into place, somewhere near that weld, you're going to have a region where the P91 is not in the proper condition. When you have a tube failure (notice, not 'if'), your repair will also add a region where the P91 is not in the proper condition.

 

[stanweld]

One of the primary problems that I saw in the early use of P91 for HRSG units is that the Engneer/Designer never specified requirements for heat treatment after hot bending as required under 129.3.3 (A) of ASME B31.1. By the way, the Designer also almost never specified heat treating requirements for hot bends in P11 and P22 alloy piping as defined in B31.1

 

[metengr]

SMF1964;
Just because a particular job is turnkey doesn't mean less involvement. EPC (engineer/procure and construct) is a new way of doing business by our procurement folks. We still insist on hold points during fabrication and construction of boiler components - new or retrofits. It would be foolish to let this fall into the hands of others that have no special interest from the customers perspective. Our company will hire outside help, if need be, to assure compliance with our tight fabrication specifications - especially when it comes to P or T91 components.

Second item is weld repair of P91 or T91 tube material. This material can be properly weld repaired provided very strict controls are followed. This requires the correct selection of welding electrodes, maintaining of preheat, interpass temperatures and a very specific post weld heat treatment procedure (time and temperature). These are items that can be reviewed and monitored as hold points. So, I do NOT agree with your gloom and doom assessment regarding repair of P91 material.

We have performed weld repairs on T91 material in the boiler due to sootblower erosion damage with no adverse affects.

 

[davefitz]

Thanks Meteng + Stanweld. Once again , you both have helped in a tough spot.

The potential problem is not as minor as a boiler tube leak- it is instead a concern related to a 14" OD main steam transfer pipe in the work space. And it is not just our site; the pipe vendor said he and all other fly- by- night shops do it the same way. This means that once they start to fail, no P91 transfer pipe will be considered safe. It will be like the Mohave P11 failure all over again.

What I am finding is that the EPC vendors ( big players , at that) and pipe shops do not have metallurgists on staff and their designers have no significant understanding of P91 metallurgy. Time for all shops to conveniently lose their recrds.

 

[stanweld]

davefitz,
It's not only on 14" lines. I've seen hot bends on 24" hot reheat lines 18" main steam lines and on down to 8" lines.

This brings on another observation. A burst in the bend of a superheater crossover line recently occurred at a Power plant that we did not construct. Elbow was supposed to have been P91 but was found to be P5 - finding a tad late! We have discovered many such mixes but PMI has almost never been required by owners or engineers. It's as if the lessons learned decades ago regarding mixed steels were never learned.